Colonic adenomatous polyps are common tumors occurring in approximately 50% of Western populations with about 10% risk of malignant progression. Key factors in the pathogenesis of colonic neoplasia are germline mutations in the adenomatous polyposis coli (ARC) gene and chronic inflammation as exemplified by patients with familial adenomatous polyposis or ulcerative colitis, respectively, who are highly susceptible to developing colon cancer. Bacteroides fragilis are common colonic commensals (up to 70% of adults colonized) and the leading anaerobe in human disease. One molecular subset of B. fragilis, enterotoxigenic B. fragilis (ETBF), is a human inflammatory diarrheal disease pathogen. However, asymptomatic colonization with ETBF is common (between 4% to 30% of individuals) and recent data link ETBF to active inflammatory bowel disease and colorectal cancer. The only known virulence factor of ETBF is a 20 kDa zinc-dependent metalloprotease toxin termed B. fragilis toxin (BFT). By cell-surface proteolysis, matrix metalloproteases regulate eukaryotic cell signal transduction and function including cell proliferation, tumor progression and metastasis. Our data reveal that BFT treatment of human intestinal epithelial cells (IEC) stimulates cleavage of the intercellular adhesion protein, E-cadherin, and triggers the activation of multiple signal transduction pathways including beta-catenin, gamma-secretase, tyrosine kinases, Nuclear Factor-kB (NF-kB) and mitogen-activated protein kinases (MARK) resulting in c-Myc induction, IEC proliferation with enhanced cellular migration and IEC secretion of the proinflammatory cytokine, interleukin-8 (IL-8). The proliferative and inflammatory potential of ETBF are further supported by our preliminary data showing ETBF colonization yields colonic hyperplasia and inflammation in conventional mice;rapid, lethal colitis in germ-free mice;and increased colon adenomas in multiple intestinal neoplasia (Min) mice. Our data show that BFT secretion is critical to murine ETBF colonic disease. Thus, we hypothesize that chronic colonization with ETBF, at least in part through BFT, acts as a co-factor in the pathogenesis of colonic neoplasia and inflammation via cleavage of E-cadherin and nuclear activation of protoooncogenes and proinflammatory genes. To test this hypothesis, we will: 1) identify the BFT receptor. We postulate this receptor mediates, in part, colon inflammation and hyperplasia stimulated by ETBF in vivo.;and 2) determine if ETBF colonization promotes colonic tumor formation in mouse models prone to colon neoplasia. Colon cancer is the third most common cancer worldwide and the second leading cause of cancer deaths in the United States in men and women. Our studies may indicate that a common stool bacterium is oncogenic and contributes to colon cancer risk. If so, these data will lead to new public health approaches to the prevention of colon cancer.